Harbours and Coastal Military Bases in Egypt in the Second Millennium B.C. Avaris, Peru-Nefer, Pi-Ramesse
Total Page:16
File Type:pdf, Size:1020Kb
Harbours and Coastal Military Bases in Egypt in the Second Millennium B.C. Avaris, Peru-nefer, Pi-Ramesse MANFRED BIETAK 1. Introduction During the Middle Kingdom, it seems that settlement in Lower Egypt was concentrated in the eastern part and at the extreme western edge of the Delta. Thus far, however, the greater part of the western half of the deltaic landscape has not yielded any sites of this period. It seems that this situation continued until the time of the New Kingdom. I have argued that the concentration of sites of the Second Intermediate period in the eastern Delta was the result of the political situation at that time and the development of a kind of homeland for an immigrated Near Eastern population which later caused the Hyksos rule in Egypt. But it now seems that this concentration of settlements in the eastern Delta had its roots in the physical geography of the Delta already in the Middle Kingdom. One has to face the fact that the western part of the Delta was void of habitation sites. It seems that only the cults at sacred places such as Sais and Buto were kept going, but we have no evidence of settlement there during this period. This is the result of extensive archaeological surveys conducted by the EES, the German Archaeological Institute, the University of Amsterdam and University of Liverpool (see Figure 1).1 Explanations for this situation have 1 See the survey of the EES: http://www.deltasurvey.ees.ac.uk/ds-home.html. At this homepage one finds a full bibliography on the survey activities in the Nile Delta. I am indebted to Alan Jeffrey Spencer for more information on this sur- vey. For the survey of the University of Amsterdam see: VAN DEN BRINK 1987; 1988. For the Survey of the University of Liverpool s. SNAPE 1986. For the survey 53 Manfred Bietak thus far not been put forward. However, it would seem that there are not many options that might explain this apparent lack of sites. One possibility is that sediment accumulation has caused sites to disappear under substantial layers of Nile mud.2 This is not very likely, however, as the subsidence rates in most parts of the western Delta are moderate,3 while sites older than the Middle Kingdom, dating to the fourth and third millennium B.C., have been found in this region.4 One has to add that already existing settlement mounds (Tells), which jut out from the floodplain and provide secure settling ground during the flood season, were usually chosen as habitation sites. Figure 1. The Delta with sites of the Middle Kingdom and the Second Intermediate Period (graphic work Marian Negrete-Martinez). As an alternative hypothesis to explain the void of Middle Kingdom and Second Intermediate Period sites in this area one could also propose that sediment accumulation rates may not have sufficiently compensated for the eustatic rise of the Mediterranean Sea level, a situation which would have led to a long- of the German Archaeological Institute s. Schiestl, https://dainst.academia.edu/ RobertSchiestl. 2 STANLEY/TOSCANO, 2009, p. 161-167. 3 According to WUNDERLICH/ANDRES, 1991, p. 115-118. 4 See note 2. 54 Harbours and Coastal Military Bases in Egypt in the Second Millennium B.C. term submersion, rendering large parts of the western Delta uninhabitable. It seems that even in the Late Period the western central Delta was less inhabited than the eastern part.5 According to the medieval chroniquer Abu el-Hassan el-Makhzoumi there was a substantial sea incursion at 961 AD6 causing the creation of the present coastal Nile Delta lakes which show remains until today in shore sediments south of the present inshore lakes.7 It is possible that the central western Delta had a repetitive weakness in respect to sea incursions. There is also the possibility of an additional effect of tsunamis after volcanic or tectonic events in the Aegean or in Asia-minor.8 With the wetlands inundated and only the levees and mounds emerging, the land would have been deprived of agriculture and flock-keeping and thus of sustenance. The reconstruction map of Stanley and Warne9 for 4000 non-calibrated years BP, based on numerous cores, shows extended wetlands reaching far south into the Delta, west of the so-called Mendesian branch of the Nile, whereas eastwards, the floodplain suitable for agriculture and pasture reached as far north as the region of what is today San el-Hagar (see Figure 2). The eastern Delta with its substantial Pleistocene sand substratum and its numerous turtlebacks (sand geziras) differs from the western Delta by offering an ideally stable settling ground.10 Additionally it seems that sediment accumulation rates were higher there than in the central and western Delta.11 The extreme western edge of the Delta must also have been a more stable place of settlement in the Middle Kingdom. Therefore, the western – and even more so the easternmost – Nile branches must have been the most suitable water courses also for harbours. 5 WILSON and GRIGOROPOULOS 2009. 6 I am grateful to PENELOPE WILSON of Durham University of informing me about this event. 7 SHAFEI, 1962; FRIHY, 1992, p. 392. 8 Some information can be gathered from Papyrus Hearst of the early Eighteenth Dynasty, which renders a magic spell about the god Seth who seems to have stopped an incursion of the Mediterranean – an event which has to be tied because of the invocation of god Seth to the eastern Delta: “Just as Seth has banned the Mediterranean Sea Seth will ban you likewise...” (GOEDICKE, 1984, p. 46). 9 STANLEY/WARNE, 1998, fig. 8D. 10 ANDRES/WUNDERLICH, 1992. 11 STANLEY et al., 1992, p. 30-39, figs. 10, 12. 55 Manfred Bietak Figure 2. Reconstruction of the Delta in the time of 4000 BP (STANLEY/WARNE, 1998, fig. 8D). 2. The Nile Regime, the harbour situation and the identification of Avaris, Peru-nefer, Pi-Ramesse Before the construction of the barrages in the nineteenth century A.D., perennial harbours for seagoing ships had to be situated within the reach of the sea.12 The period of drought between March and early July, when the Nile shrank to one fifth of its average volume, made river navigation very difficult.13 During this period, seawaters penetrated the nearly empty channels of the Delta and made perennial navigation within the reach of the sea possible. For this reason we find deltaic harbours such as Rashid (Rosette) and Dumiat (Damiette) in locations protected against winter gales, yet within a relatively short distance to the sea until the nineteenth century. Also Tanis served as a harbour for seagoing traffic at least until the fifth century A.D.14 Harbours further upstream, beyond the reach 12 BIETAK, 2010a, p. 165-169; 2010b, p. 19-21. 13 LE PÈRE, 1822, p. 140-141; CLOT, 1840; COOPER, 2012. 14 A manuscript in the monastery of Arezzo gives an account of a pilgrimage by ab- bess Aetheria to the Holy Land. The ship which she uses for her transfer to Egypt landed at Tanis from where the travel continued overland from the Nile Delta to 56 Harbours and Coastal Military Bases in Egypt in the Second Millennium B.C. of the sea during the months of low river levels, could not provide perennial navigation for sea-bound traffic and had difficulties to reach the Mediterranean for nearly half of the year. On top of such limitations we know from Aramaic custom duty papyri with lists of incoming and outgoing ships from the Persian Period from Elephantine that there was no sea traffic at all during the months Thoth and Paophi (approx. January and February),15 presumably because of usual fogs and winter gales during this time. Based on these environmental conditions we can once and for all exclude Memphis as a candidate for the identification with the famous New Kingdom naval base of Peru-nefer16. It is unthinkable that the major naval base targeting the Near East was only operational from the second half of July until the end of the year during times of increased Egyptian warfare; and even then, the long distance from the Mediterranean would prolong the reaction of the Egyptian crown to any happenings in the Levant or would render a necessary mission impossible for six months. Therefore, we have to look for an alternative candidate, which was found in a huge harbour basin at Tell el-Dab‘a connected with entry and exit channels to the Nile system within reach of the sea in the second millennium the Sinai, to the Holy Land and back. See BROX, 1995; RÖWEKAMP, 1995; VRETSKA, 1958. 15 PORTEN/YARDENI, 1993, C3.7; YARDENI, 1994, p. 69. The name of the harbour is not mentioned, but it is clearly a harbour for seagoing ships (YARDENI, 1994, p.77, n. 13). 16 BADAWI, 1948, p. 34-36, 55-63, 137-139; GLANVILLE, 1931, p. 109; ID., 1932; HELCK, 1939, p. 49-50; ID., 1971, p. 160, 166, 356-357, 447-448, 454-456, 460, 471, 473, 501; JEFFREYS/SMITH, 1988, p. 61; EDEL, 1977, p. 155; KAMISH, 1985; EAD., 1986; DER MANUELIAN, 1987, p. 159; SÄVE-SÖDERBERGH, 1946, p. 37–39; STADELMANN, 1967, p. 32-35; ZIVIE, 1988, p. 107. Recently FORSTNER-MÜLLER, 2014, argued again in favour of locating Peru-nefer at Memphis and suggested that this town was reachable during the time of inundation, which lasted only a few months and is therefore unfeasible. The fact that she did not find any New Kingdom remains before the Ramesside Period in the very restricted area she excavated at the northern edge of the big harbour basin at Tell el-Dab‘a is not a valid argument as we don’t know yet where the Eighteenth Dynasty installations were positioned.